- **Main CFD Forum**
(*http://www.cfd-online.com/Forums/main/*)

- - **fundamental questions
**
(*http://www.cfd-online.com/Forums/main/8905-fundamental-questions.html*)

fundamental questions
Dear CFD users,
I have two questions about CFD. 1. In general, why is orthogonal mesh preferabble rather than triangular mesh? 2. Can FVM solver solve fluid-structure interaction (FSI) problems? or can only FEM solver solve FSI problems? I'd appreciate a lot for you answers in advance. |

Re: fundamental questions
Hi Jim
For orthogonal meshes: 1) Orthogonal mesh is preferabble in cases where you can align the mesh with the flow direction (face normal of a CV = flow direction through the CV). If the flow is not aligned with the face normals, then you can induce errors in the calculation. See Patankar's book on the error induced due to non-alignment of mesh. 2) In FV formulation for an unstructrued mesh, typically the defect correction approach is used for the diffusion terms. This requires an approximation for the gradient at the face of a CV. For highly non-orthogonal meshes, sometimes, this approximation becomes very bad and the solution convergence is significantly hampered. There are probably more reasons which other users can site. For FSI+FVM: 1) Yes FVM solvers can solve FSI+FVM |

All times are GMT -4. The time now is 22:29. |